Identifying functional gene regulatory network phenotypes underlying single cell transcriptional variability.

Area of Science:

• Neuroscience
• Computational Biology
• Systems Biology

Background:

• Single-cell transcriptomic data reveals neuronal subtypes organized by gene expression gradients.
• Understanding the regulatory networks driving these transcriptional states is crucial but remains unclear.

Purpose of the Study:

• To develop a novel fuzzy logic-based approach for inferring quantitative gene regulatory network models from single-cell gene expression data.
• To identify causal gene interactions and regulatory network dynamics in neuronal subtypes.

Main Methods:

• Developed an a priori regulatory network and trained it using in vivo single-cell gene expression data.
• Employed a fuzzy logic-based approach to infer quantitative gene regulatory network models.
• Simulated network responses to experimentally observed stimuli levels.

Main Results:

• The inferred gene regulatory network accurately mirrored observed gene expression patterns and quantitative ranges in individual neurons.
• Distinct regulatory interactions and stimuli drive variable gene expression across neuronal subtypes.
• Identified a lack of negative feedback regulation in the catecholaminergic subtype network and demonstrated how distinct stimuli can drive divergent subtypes to similar states.

Conclusions:

• Heterogeneous single-cell gene expression profiles should be analyzed using regulatory network modeling.
• This approach separates contributions from network interactions versus cellular inputs.
• Reveals subtype-specific regulatory differences and potential for state convergence.